EP1760590B1

EP1760590B1 - Storage controller system having a splitting command for paired volumes and method therefor

Info

Publication number: EP1760590B1
Application number: EP06024150A
Authority: EP
Inventors: Susumu c/o Hitachi Ltd. Intel. Prop. Grp. Suzuki; Masanori c/o Hitachi Ltd. Intel. Prop. Grp. Nagaya; Takao c/o Hitachi Ltd. Intel. Prop. Grp. Sato
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2002-12-18
Filing date: 2003-10-01
Publication date: 2007-12-26
Anticipated expiration: 2023-10-01
Also published as: EP1956487A1; EP1956487B1; US20050251636A1; EP1431876A3; ATE443290T1; JP4704660B2; EP1760590A1; US7962712B2; US7334097B2; JP2004199336A; EP1431876A2; US20060168412A1; US20080288733A1; ATE400021T1; DE60329341D1; US7418563B2; US20040133752A1; US7089386B2; DE60318337T2; US20070028066A1

Abstract

The invention provides a storage device controller (200) adapted to be connected to a plurality of storage devices (300) provided with a plurality of storage volumes for storing data, said storage device controller being connectable to an information processing apparatus (100) for sending write requests to the storage volumes, wherein the storage device controller (200) comprises means for forming pairs of storage volumes, each pair bringing one of the plurality of storage volumes into correspondence with another storage volume. The storage device controller (200) is adapted to start to split pairs of a set of paired volumes upon receiving a split command for the set of paired volumes, wherein the storage device controller (200) is adapted to execute a write request being received during splitting the pairs of the set of paired volumes and being addressed to a volume in the set of paired volumes, after a completion of splitting a pair related to the volume in the set of paired volumes.

Description

The present invention relates to a method for controlling a storage device controller, a storage device controller, and a storage system.

2. Description of the Related Arts

There is a well-known copy management function used in a storage system that includes an information processing apparatus and a disk array unit connected to each other for communications. The function manages primary volume data in duplicate by copying data from a primary volume to a secondary volume in real time. The primary (master) volume that is a source of copy and the secondary (sub) volume that is a destination of copy are paired.
In such a storage system, however, data often overflows one primary volume into other primary volumes during communications between the information processing apparatus and the disk array unit. If an attempt is made to back up the data in such an occasion, a plurality of pairs (of primary and secondary volumes) must be reset from the paired state. If data in a primary volume for which the pair is already reset is updated during sequential resetting of paired states, the data is not updated in its corresponding secondary volume while data in a primary volume of which pair state is not reset is updated in its corresponding secondary volume sometimes.
US 5,692,155 discloses a data storage system which atomically suspends multiple duplex pairs across either a single storage subsystem or multiple storage subsystems. The duplex pairs are suspended such that the data on the secondary DASDs of the duplex pairs is maintained in a sequence consistent order. A host processor in the data storage system running an application generates records and record updates to be written to the primary DASDs of the duplex pairs. The storage controller directs copies of the records and record updates to the secondary DASDs of the duplex pairs. Sequence consistency is maintained on the secondary DASDs by quiescing the duplex pairs and then suspending the duplex pairs with change recording. Quiescing the duplex pairs allows any current write I/O in progress to complete to the primary DASD. The storage controller then locks out any subsequent write I/O from the host processor by raising a long busy signal to such subsequent write requests.
US 6,301,643 disclosed a system for maintaining consistency of data across storage devices. A cut-off time value is provided to the system. The system then obtains information on data writes to a first storage device, including information on time stamp values associated with the data writes indicating an order of the data writes to the first storage device. At least one group of data writes having time stamp values earlier in time than the cut-off time value is then formed. The system then transfers the data writes in the groups to a second storage device for storage therein.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide a method for controlling a storage, device controller, a storage device controller, and a storage system capable of managing copies of data while keeping the consistency among the data stored in a plurality of storage volumes.
These objects are accomplished by a storage system according to claim 1 or 6, a storage device controller according to claim 11 and a method for controlling a storage device controller according to claim 12.
The storage device and the storage device controller may be included in a disk array unit. The information processing apparatus and the disk array unit are included in the storage system.
Storage volumes are storage resources provided in the disk array unit or storage device and they are divided into physical volumes and logical volumes. A physical volume is a physical storage area provided in a disk drive of the disk array unit or storage device and a logical volume is a storage area allocated logically in a physical volume.
The "paired" means a state in which two storage volumes are brought into correspondence with each other as described above.

BRIEF DESCRIPTION OF THE DRAWINGS

Preferred embodiments of the present invention will now be described in conjunction with the accompanying drawings, in which:

Fig.1 is an overall block diagram of a storage system in an embodiment of the present invention;
Fig.2 is a block diagram of an information processing apparatus in the embodiment of the present invention;
Fig.3 is a block diagram of a channel adapter provided in a storage device controller in the embodiment of the present invention;
Fig.4 is a table stored in a shared storage provided in the storage device controller in the embodiment of the present invention;
Fig.5 is pairs of storage volumes in the embodiment of the present invention;
Fig.6 is a flowchart of the processings of the storage device controller for splitting a pair for explaining the present invention;
Fig.7 is a flowchart of the processings of the storage device controller for splitting a pair and inputting/outputting the split pair data items for explaining the present invention;
Fig.8 is a table stored in the shared storage provided in the storage device controller in the embodiment of the present invention; and
Fig.9 is a flowchart of the processings of the storage device controller for splitting a pair.

DETAILED DESCRIPTION OF THE INVENTION

Hereunder, the preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

=== overall Configuration ===

At first, the storage system in an embodiment of the present invention will be described with reference to the block diagram shown in Fig.1.
An information processing apparatus 100 is a computer provided with a CPU (Central Processing Unit), a memory, etc. The CPU of the information processing apparatus 100 executes various types of programs to realize various functions of the apparatus 100. The information processing apparatus 100 is used, for example, as a core computer in an automatic teller machine in a bank, a flight ticket reservation system, or the like.
The information processing apparatus 100 is connected to a storage device controller 200 to communicate with the controller 200. The information processing apparatus 100 issues data input/output commands (requests) to the storage device controller 200 to read/write data from/to the storage devices 300. The information processing apparatus 100 also sends/receives various commands to/from the storage device controller 200 to manage the storage devices 300. For example, the commands are used for managing copies of data stored in the storage volumes provided in the storage devices 300.
Fig.2 shows a block diagram of the information processing apparatus 100.
The information processing apparatus 100 is configured by a CPU 110, a memory 120, a port 130, a media reader 140, an input device 150, and an output device 160.
The CPU 110 controls the whole information processing apparatus 100 and executes the programs stored in the memory 120 to realize various functions of the apparatus 100. The media reader 140 reads programs and data recorded on the recording medium 170. The memory 120 stores the programs and data read by the reader 140. Consequently, the media reader 170 can be used to read a storage device management program 121 and an application program 122 recorded in the medium 170 and store them in the memory 120. The recording medium 170 may be any of flexible disks, CD-ROM disks, semiconductor memories, etc. The media reader 140 may also be built in the information processing apparatus 100 or provided as an external device. The input device 150 is used by the operator to input data addressed to the information processing apparatus 100. The input device 150 may be any of keyboards, mice, etc. The output device 160 outputs information to external. The output device 160 may be any of displays, printers, etc. The port 130 is used to communicate with the storage device controller 200. In that connection, the storage device management program 121 and the application program 122 may be received from another information processing apparatus 100 through the port 130 and stored in the memory 120.
The storage device management program 121 manages copies of data stored in the storage volumes provided in the storage devices 300. The storage device controller 200 manages copies of data with use of various copy management commands received from the information processing apparatus 100.
The application program 122 realizes the functions of the information processing apparatus 100. For example, the program 122 realizes functions of an automatic teller machine of a bank and functions of a flight ticket reservation system as described above.
Next, the storage device controller 200 will be described with reference to Fig.1 again. The storage device controller 200 controls the storage devices 300 according to the commands received from the information processing apparatus 100. For example, when receiving a data input/output request from the information processing apparatus 100, the storage device controller 200 inputs/outputs data to/from a storage volume provided in a storage device 300.
The storage device controller 200 is configured by a channel adapter 210, a cache memory 220, a shared storage 230, a disk adapter 240, a management terminal (SVP: Service Processor) 260, and a connection unit 250.
The channel adapter 210 provided with a communication interface with the information processing apparatus 100 exchanges data input/output commands, etc. with the information processing apparatus 100.
Fig.3 shows a block diagram of the channel adapter 210.
The channel adapter 210 is configured by a CPU 211, a cache memory 212, a control memory 213, a port 215, and a bus 216.
The CPU 211 controls the whole channel adapter 210 by executing a control program 214 stored in the control memory 213. The control program 214 stored in the control memory 213 thus enables data copies to be managed in this embodiment. The cache memory 212 stores data, commands, etc. to be exchanged with the information processing apparatus 100 temporarily. The port 215 is a communication interface used for the communication with the information processing apparatus 100 and other devices provided in the storage device controller 200. The bus 216 enables the mutual connection among those devices.
Return to Fig.1 again. The cache memory 220 stores data to be exchanged between the channel adapter 210 and the disk adapter 240 temporarily. In other words, if the channel adapter 210 receives a write command as a data input/output command from the information processing apparatus 100, the channel adapter 210 writes the command in the shared storage 230 and the target data received from the information processing apparatus 100 in the cache memory 220 respectively. The disk adapter 240 then reads the target data from the cache memory 220 according to the write command written in the shared storage and writes the read data in a storage device 300.
The management terminal 260 is a kind of information processing apparatus used for the maintenance/management of the storage device controller 200 and the storage devices 300. For example, the management terminal 260 changes the control program 214 executed in the channel adapter 210 to another. The management terminal 260 may be built in the storage device controller 200 or may be separated. The management terminal 260 may also be dedicated to the maintenance/management of the storage device controller 200 and the storage devices 300 or may be configured as a general information processing apparatus for maintenance/management. The configuration of the management terminal 260 is the same as that of the information processing apparatus 100 shown in Fig.2. Concretely, the management terminal 260 is configured by a CPU 110, a memory 120, a port 130, a recording media reader 140, an input device 150, and an output device 160. Consequently, the control program to be executed in the channel adapter 210 may be read from the recording medium 170 through the media reader 140 of the management terminal 260 or received from the information processing apparatus 100 connected thereto through the port 130 of the management terminal 260.
The disk adapter 240 controls the storage devices 300 according to the commands received from the channel adapter 210.
Each of the storage devices 300 is provided with a storage volume to be used by the information processing apparatus 100. Storage volumes are storage resources provided in the storage devices 300 and divided into physical volumes that are physical storage areas provided in disk drives of the storage devices 300 and logical volumes that are storage areas allocated logically in those physical volumes. The disk drives may be any of, for example, hard disk drives, flexible disk drives, semiconductor storage devices, etc. The disk adapter 240 and each of the storage devices 300 may be connected to each other directly as shown in Fig.1 or through a network. The storage devices 300 may also be united with the storage device controller 200 into one.
The shared storage 230 can be accessed from both of the channel adapter 210 and the disk adapter 240. The shared storage is used to receive/send data input/output requests/commands and store management information, etc. of the storage device controller 200 and the storage devices 300. In this embodiment, the shared storage 230 stores a consistency group management table 231 and a pair management table 232 as shown in Fig.4.

=== Pair Management Table ===

The pair management table 232 is used to manage copies of data stored in the storage devices 300. The table 232 has columns of "pair", "primary volume", "sub volume", "pair state", and "consistency group".
The "pair" column holds pair names. A pair means a combination of two storage volumes. Fig.5 shows an example of paired storage volumes. In Fig.5, two pairs, that is, pairs A and B are denoted. One of paired volumes and the other of the paired volumes are managed as a primary volume and a secondary volume. In Fig.5, a primary volume is described as a master volume and a secondary volume is described as a sub volume. A plurality of secondary volumes can be combined with one primary volume.
Return to the pair management table 232 shown in Fig.4. The "primary" column describes primary volumes paired with secondary volumes while the "secondary" column describes secondary volumes paired with primary volumes.
The "pair state" column describes the state of each pair of volumes. The "pair state" is classified into "paired", "split", and "re-sync".
The "paired" denotes that data in a secondary volume is updated with the data in its corresponding primary volume written by the information processing apparatus 100. The consistency of the data stored in a pair of primary and secondary volumes is assured with such correspondence set between those primary and secondary volumes.
The "split" denotes that data in a secondary volume is not updated with the data in its corresponding primary volume written by the information processing apparatus 100. Concretely, while primary and secondary volumes are in such a "split" state, the correspondence between those volumes is reset. Consequently, the data consistency is not assured between those primary and secondary volumes. However, because data in any secondary volume that is in the "split" state is not updated, the data in secondary volumes can be backed up during the while; for example, data stored in secondary volumes can be saved in a magnetic tape or the like. This makes it possible to back up data while the data in primary volumes is used continuously during the backup operation for a job that has been executed by the information processing apparatus 100.
The "re-sync" denotes a transition state of a pair of volumes, for example, from "split" to "paired". More concretely, the "re-sync" means a state in which data in a secondary volume is being updated with the data written in its corresponding primary volume while the pair is in the "split" state. When the data in the secondary volume is updated, the state of the pair is changed to "paired".
To form a pair of storage volumes or to change the state of the pair from "paired"/"split" to "split"/"paired", the operator instructs the information processing apparatus 100 in which the storage device management program 121 is executed through the input device 150. A command from the operator is then sent to the channel adapter 210 of the storage device controller 200. After that, the channel adapter 210 executes the control program 214 to form a pair of storage volumes or change the state of the pair according to the command. According to the state of the formed pair of storage volumes, the channel adapter 210 controls the object storage volumes, for example, updating a secondary volume with a copy of data updated in its corresponding primary volume when those volumes are "paired".
As described above, the channel adapter 210 changes the states of pairs one by one sequentially. This is because one primary volume can be paired with a plurality of secondary volumes as described above and if the states of a plurality of pairs are changed simultaneously, the management of primary volumes comes to become complicated.
Forming a pair of volumes and changing the state of each pair of volumes can also be made automatically at a predetermined time or according to a command received from another information processing apparatus 100 connected through the port 130 independently of instructions from the operator.

=== Consistency Group ===

The "consistency group" column describes the number of each consistency group (pair group) consisting of pairs of volumes. A consistency group means a group of a plurality of storage volume pairs to be controlled so that the states of those pairs are changed to the "split" together. Concretely, a plurality of pairs in a consistency group are controlled so that their states are changed to the "split" simultaneously (hereinafter, this processing will be referred to as the synchronism among the state changes to the "split") while the states of a plurality of paired volumes are changed one by one sequentially as described above.
For example, assume now that the information processing apparatus 100 writes data in a storage volume while the pair states of a plurality of paired volumes in a consistency group are changed sequentially from "paired" to "split". If no consistency group is formed and the data is written in a paired primary volume after the pair state is changed to the "split", the data is not written in its corresponding secondary volume. If the data is written in a paired primary volume of which state is not changed to the "split" yet, the data is also written in the secondary volume. If the paired primary volume belongs to a consistency group at that time, however, the data is not written in its corresponding secondary volume regardless of the pair state of the primary volume (whether it is in the "split" or not). This is because the data is written in the primary volume after pair splitting (resetting of the correspondence between primary and secondary volumes) is started in the consistency group.
Forming a consistency group with a plurality of pairs such way is effective for a case in which data is to be stored in a plurality of storage volumes, for example, when write data is too large to be stored in one storage volume and when it is controlled so that one file data is stored in a plurality of storage volumes.
Such assured synchronism of the pair state changes of volumes to the "split" in a consistency group is also effective for writing/reading of data in/from secondary volumes requested from the information processing apparatus 100.
Concretely, if no consistency group is already formed, data can be written/read in/from any paired secondary volume after the pair state is changed to the "split" while it is inhibited to write/read data in/from any secondary volume of which pair state is not changed to the "split".
A batch split receiving flag (ID information) of the consistency group management table 231 is used to assure the synchronism of such pair state changes of volumes to the "split" in the above consistency group. Next, the processings for assuring such synchronism which are possible generally will be described with reference to the flowchart shown in Fig.6 for explaining the invention.

=== Processing Flow ===

The following processings are executed by the CPU 211 provided in the channel adapter 210 with use of the control program 214 (program) consisting of codes for realizing various operations.
At first, the channel adapter 210 receives a pair splitting request (split command) addressed to a consistency group from the information processing apparatus 100(S1000). The channel adapter 210 then turns on the batch split receiving flag in the consistency group management table 231 stored in the shared storage 230 (S1001). After that, the channel adapter 210 begins to change the pair state of a not-split pair of volumes in the consistency group to the "split" (S1003). Concretely, the channel adapter 210 resets the correspondence between the primary volume and the secondary volume in the pair and stops updating of the data in the secondary volume with the data written in the primary volume. The channel adapter 210 then changes the description for the pair in the "paired" column in the pair management table 232 to "split" (S1004). Those processings are repeated for each pair in the consistency group. When the states of all the pairs in the consistency group are changed to the "split" (S1005), the channel adapter 210 turns off the batch split flag, then exits the processing.
If the channel adapter 210 receives a read/write request from the information processing apparatus 100 during the above processing, the adapter 210 checks whether or not the request is addressed to a not-split storage volume, that is, a "paired" storage volume (for which the correspondence to its secondary volume is not reset)(S1006). If the check result is YES (addressed), the adapter 210 changes the pair state of the volume to the "split" (S1007). The adapter 210 then changes the description of the pair in the pair state column in the pair management table 232 to the "split" (S1008) and executes the data read/write processing (input/output processing)(S1009).
On the other hand, if the check result in (S1006) is NO (not addressed), this means that the command is addressed to a "split" volume. The adapter 210 thus executes the read/write processing for the storage volume (S1009) immediately.
Consequently, the synchronism of the pair state changes of "paired" volumes to the "split" in a consistency group is assured.
In the flowchart shown in Fig.6, if the channel adapter 210 receives a read/write request from the information processing apparatus 100 while splitting paired volumes in a consistency group sequentially, the adapter 210 checks whether or not the request is addressed to a not-split pair of volumes (S1006) to execute the read/write processing (S1009). However, according to the claimed invention the adapter 210 suppresses the execution of the read/write processing requested from the information processing apparatus 100 while the adapter 210 splits paired volumes in a consistency group sequentially. In that connection, the adapter 210 can execute the read/write processing after the adapter 210 completes splitting of all the paired volumes in the consistency group and turns off the batch split flag.
Fig.7 shows a flowchart for those processings by the channel adapter 210 in detail.
At first, the channel adapter 210 forms a consistency group for both pairs A and B according to a command received from the information processing apparatus 100(S2000 to S2002). The command is inputted, for example, by the operator through the input device 150 of the information processing apparatus 100. The command, inputted to the information processing apparatus 100 is sent to the channel adapter 210 by the storage device management program 121. The "paircreate -g GRP0" shown in Fig.7 is such a command. Receiving the command, the channel adapter 210 forms a consistency group, then records predetermined data in the pair management table 232 and the consistency group management table 231 stored in the shared storage 230 respectively. Fig.4 shows how the predetermined data is recorded in those tables 231 and 232. However, although the state of the pair A is described as "split" in the pair state column in the pair management table 232 shown in Fig.4, the state of the pair A at that time is actually "paired". Similarly, although "ON" is described in the batch split receiving flag column for the consistency group 0 in the consistency group management table 231, the actual state at that time is actually "OFF".
The channel adapter 210, when receiving a read/write request (R/W1) for the storage volume 1 in the pair A from the information processing apparatus 100(S2008), executes the read/write processing as usually (S2009). This is because "OFF" is described in the batch split receiving flag column for the consistency group 0 in the consistency group management table 231.
After that, the information processing apparatus 100 instructs the channel adapter 210 to split the pair B in the consistency group 0 with a command (S2003). The "pairsplit -g GRP0" shown in Fig.7 is an example of the command issued at that time. This command may also be inputted by the operator through the input device 150 of the information processing apparatus 100.
The channel adapter 210 then turns on the batch split receiving flag for the consistency group 0 in the consistency group management table 231 stored in the shared storage 230 (S2004) to start splitting of each pair sequentially (S2005, S2006). Fig.4 shows the pair management table 232 in which the pair A is split. Completing splitting of all the target pairs, the channel adapter 210 turns OFF the batch split receiving flag and exits the processing (S2007).
If the channel adapter 210 receives a read/write request (R/W2) addressed to the storage volume 3 of the pair B from the information processing apparatus 100 (S2010) while the channel adapter 210 turns ON the batch split receiving flag (S2004) after receiving a split command addressed to the consistency group 0 from the information processing apparatus 100, the channel adapter 210 executes the read/write processing as usually (S2011). This is because "OFF" is still set in the batch split receiving column for the consistency group 0 in the consistency group management table 231.
However, if the channel adapter 210 receives a read/write request (R/W3) addressed to the storage volume 3 of the pair B from the information processing apparatus 100 (S2012) after turning ON the batch split receiving flag (S2004), the channel adapter 210 splits the pair B (S2013), then executes the read/write processing (S2014).
As described above, the channel adapter 210, when receiving a read/write request from the information processing apparatus 100, refers to the batch split receiving flag to check whether or not it is after resetting of the pair state of each pair in the consistency group is started that the read/write command has been issued.
If the channel adapter 210 receives the read/write request (R/W4) after completing splitting of the pair A in (S2005), the channel adapter 210 executes the read/write processing (S2016). This is because "split" is set for the pair A in the pairing column in the pair management table 232 and the channel adapter 210 knows that "split" denotes that the pair A is split.
In that connection, no splitting processing is done for the pair B in (S2005), since the pair B is already split during the read/write processing in (S2013).
Because the batch split receiving flag is provided as described above, the synchronism among the pair state changes of all the pairs in a consistency group to the "split" is assured.

=== Consistency Group Management Table ===

Next, a description will be made for another embodiment of the present invention with respect to the management information in the consistency group management table 231.
In this embodiment, each split starting time is recorded in the consistency group management table 231 as shown in Fig.8. In the example shown in Fig.8, splitting of pairs in the consistency group 0 is started at 12:00. When splitting of all the pairs in the consistency group 0 is completed, the description in the split starting time column is changed to "-".
A split starting time is specified with a command received from the information processing apparatus 100. Such split starting may also be specified so as to be started immediately with a command; no concrete time is specified in such an occasion. In that connection, the current time is recorded in the split starting time column.
In this embodiment, the channel adapter 210, when receiving a read/write command from the information processing apparatus 100, compares the read/write command issued time recorded in the read/write command (request) with the time described in the split starting time column of the consistency group management table 231. If the command issued time is later, the channel adapter 210 executes the read/write processing after the end of the splitting.
This is why it is possible to assure the synchronism among the state changes of pairs in a consistency group to the "split".

=== Processing Flow ===

Next, how the above processings are possibly executed will be described in detail with reference to the flowchart shown in Fig.9 for explaining the invention.
The processings are executed by the CPU 211 of the channel adapter 210 with use of the control program 214 consisting of codes for realizing various operations
At first, the channel adapter 210 receives a pair splitting request (split command) addressed to a consistency group from the information processing apparatus 100 (S3000). The channel adapter 210 then records the split starting time recorded in the split command in the split starting time column of the consistency group management table 231 stored in the shared storage 230 (S3001). After that, the channel adapter 210 compares the split starting time with the current time to check whether or not the split starting time is passed (S3003). If the check result is YES (passed), the channel adapter 210 begins the state change of a not-split pair in the consistency group to the "split" (S3004). Concretely, the channel adapter 210 resets the correspondence between primary and secondary volumes of the pair and suppresses updating of the data in the secondary volume with the data written in the primary volume. The channel adapter 210 then changes the description for the pair in the pair state column in the pair management table 232 to "split" (S3005). The above processings are repeated for all of the pairs in the consistency group. When the states of all the pairs in the consistency group are changed to "split" (S3006), the channel adapter 210 changes the description for the pair in the split starting time column to "-" and exits the processing (S3007).
If the channel adapter 210 receives a read/write request from the information processing apparatus 100 during the above processing, the channel adapter 210 checks whether or not the request is addressed to a not-split pair, that is, a "paired" storage volume (the correspondence is not reset)(S3008). If the check result is YES (addressed), the channel adapter 210 compares the command issued time recorded in the command with the split starting time (S3010). If the command issued time is later, the channel adapter 210 changes the pair state to the "split" (S3011), then changes the description for the pair in the pair state column in the pair management table 232 to "split" (S3012). After that, the channel adapter 210 executes the read/write processing (input/output processing) (S3013).
On the other hand, if the read/write command is addressed to a split pair in (S3008), that is, a "split" storage volume or the command issued time recorded in the request is earlier than the split starting time, the channel adapter 210 reads/writes data from/in the storage volume (S3009).
This is why it is possible to assure the synchronism among the state changes of the pairs in a consistency group to the "split".
In the flowchart shown in Fig.9, if the channel adapter 210 receives a read/write request from the information processing apparatus 100 while splitting pairs in a consistency group sequentially, the channel adapter 210 checks whether or not the request is addressed to a not-split storage volume (S3008) and executes the read/write processing (S3009, S3013). However, according to the invention the channel adapter 210 suppresses execution of the read/write processing even when receiving a read/write request from the information processing apparatus 100 while splitting pairs in a consistency group sequentially as described above. The channel adapter 210 executes the read/write processing after completing splitting of all the pairs in the consistency group and changing the description for the pair in the split starting time column to "-".
In this embodiment, consistency groups are formed by storage devices 300 connected to the same storage device controller respectively. However, the present invention is not "limited only to that embodiment. In this embodiment, consistency groups should preferably be formed by storage devices 300 connected to a plurality of storage device controllers respectively. In that connection, a consistency group may be formed over a plurality of storage device controllers 200 that come to communicate with each another to create the consistency group management table 231 and the pair management table 232. The consistency group management table 231 and the pair management table 232 may be managed by one of the storage device controllers 200 and shared with other storage device controllers 200 or each of those storage device controllers manages the same table. Furthermore, volumes controlled by a plurality of storage device controllers 200 should preferably be paired in this embodiment. In that connection, a pair might be formed over a plurality of storage device controllers 200 and those storage device controllers 200 come to communicate with each another to create the consistency group management table 231 and the pair management table 232. In that connection, the consistency group management table 231 and the pair management table 232 may be managed by one of the storage device controllers 200 and shared with other storage device controllers 200 or those storage device controllers manages the same table respectively.
While the embodiments of the present invention have been described, the description is just for illustrative purposes, and it is to be understood that changes and variations may be made without departing from the scope of the following claims.

Claims

A storage system comprising:
a storage device controller (200) coupled to an information processing apparatus (100); and

a plurality of storage devices (300) coupled to the storage device controller (200);
wherein the storage device controller (200) forms a plurality of volumes using the plurality of storage devices (300), and forms a plurality of pairs using the plurality of volumes,
characterized in that the storage device controller (200) starts to split pairs of a set of paired volumes upon receiving a split command, created by the information processing apparatus (100), for the set of paired volumes,
wherein, after receipt of the split command, the storage device controller (200) suppresses an execution of a write request, addressed to a volume in the set of paired volumes, sent from the information processing apparatus (100), until a completion of splitting the pairs of the set of paired volumes.
A storage system according to claim 1, characterized in that the storage device controller (200) executes a write request, sent from the information processing apparatus (100), that is not addressed to a volume in the set of paired volumes during splitting pairs in the set of paired volumes.
A storage system according to claim 1 or 2, characterized in that the storage device controller (200) executes the suppressed write request addressed to the volume in the set of paired volumes after the completion of splitting pairs in the set of paired volumes.
A storage system according to at least one of the preceding claims, characterized in that the pairs of the set of paired volumes are formed by primary volumes to be addressed from the information processing apparatus (100) and secondary volumes to be updated with a copy data of a write data to be written to the primary volumes, respectively.
A storage system according to at least one of the preceding claims, characterized in that the storage device controller (200) sequentially splits the pairs in the set of paired volumes.
A storage system including a plurality of storage device controllers (200) and a plurality of storage devices (300) comprising:
a first storage device controller (200) of the plurality of storage device controllers (200) coupled to an information processing apparatus (100) configuring a plurality of first volumes; and

a second storage device controller (200) of the plurality of storage device controllers (200) configuring a plurality of second volumes;
characterized in that the first and second storage device controllers (200) form a consistency group with the first volumes and the second volumes,
wherein the first storage device controller (200) starts to split paired volumes in the consistency group upon receiving a group split command, created by the information processing apparatus (100), for the consistency group,
wherein, after receipt of the group split command, the first storage device controller (200) suppresses an execution of a write request, addressed to a volume in the consistency group, sent from the information processing apparatus (100), until a completion of splitting paired volumes in the consistency group.
A system according to claim 6, wherein the first storage device controller (200) executes a write request that is not addressed to a volume in the consistency group during splitting the paired volumes in the consistency group.
A system according to at least one of the claims 6 to 7, wherein the first storage device controller (200) executes the suppressed write request addressed to the volume in the consistency group after the completion of splitting paired volumes in the consistency group.
A system according to at least one of the claims 6 to 8, wherein the paired volumes in the consistency group are formed by primary volumes to be addressed from the information processing apparatus (100) and secondary volumes to be updated with a copy data of a write data to be written to the primary volumes respectively.
A system according to at least one of the claims 6 to 9, wherein the first storage device controller (200) sequentially splits the paired volumes in the consistency group.
A storage device controller (200) connected to a plurality of storage devices (300) provided with a plurality of storage volumes for storing data, said storage device controller being connectable to an information processing apparatus (100) for sending write requests to the storage volumes, wherein the storage device controller (200) comprises means for forming pairs of storage volumes for bringing one of the plurality of storage volumes into correspondence with another storage volume,
characterized in that
the storage device controller (200) starts to split pairs of a set of paired volumes upon receiving a split command for the set of paired volumes,
wherein the storage device controller (200) suppresses an execution of a write request addressed to a volume in the set of paired volumes until a completion of splitting the pairs of the set of paired volumes.
A method for controlling of at least one storage device controller (200) comprising the step of creating a plurality of pairs of volumes using a plurality of volumes;
characterized by the steps of:
forming a consistency group by selecting at least two pairs based upon a group create command from an information processing apparatus (100);

splitting the at least two pairs in the consistency group after receiving a group split command for the consistency group, the command being created by the information processing apparatus (100); and

suppressing, after receipt of the group split command, an execution of a write request addressed to a volume of the at least two pairs belonging to the consistency group until completion of splitting the at least two pairs.
A method according to claim 12, further comprising the step of performing a write request addressed to a volume that is not in the consistency group before completion of splitting the at least two pairs in the consistency group.
A method according to claim 12 or 13, characterized by the step of executing the suppressed write request addressed to the volume in the consistency group after the completion of splitting the at least two pairs in the consistency group.
A method according to at least one of the claims 12 to 14, wherein the at least two pairs in the consistency group are formed by primary volumes to be addressed from the information processing apparatus (100) and secondary volumes to be updated with a copy data of a write data to be written to the primary volumes, respectively.
A method according to at least one of the claims 12 to 15, wherein the at least two pairs of volumes in the consistency group are sequentially split.